Pharmaceutical Prices:

What’s Missing in the Public Discussion

LETTER FROM THE EDITORS

The topic of pharmaceutical prices has again become a contentious one, and it is the subject of this third issue of Economic Realities in Health Care Policy.

Much of the discussion around this topic confuses "prices" with "expenditures" and fails to note that the primary driver of rising pharmaceutical spending is not rising prices. Instead, increased spending is being driven by the increased use of both familiar and innovative products. Indeed, increased spending signals that consumers place value on the improved health provided by pharmaceuticals.

There is also clamor that "very low" pharmaceutical prices in Mexico or Canada are a testimonial to "high" US prices. In fact, the prices of essentially all products vary from place to place. Pharmaceuticals are no exception. Pharmaceutical prices differ both from store to store within local markets and from one country to another. Price differences exist for a whole host of reasons that lie beyond a pharmaceutical manufacturer’s control––such as changes in exchange rates and differences in retailers’ margins, to name but two.

We are optimistic that the information provided in this issue will promote greater understanding of pharmaceutical prices and provide greater balance to discussions of this issue.

Future issues will address the importance of consumer empowerment and direct-to-consumer advertising, as well as additional aspects of the health/workplace productivity link, the topic with which we started the series.

As always, we appreciate the many comments we have received from you and invite your thoughts on this and other topics of importance in developing sound health care policies.

Alison Keith, PhD
Richard L Manning, PhD
Pfizer Inc.

Pharmaceutical Expenditures Are Rising...
Driven by Value, Not Prices

Pharmaceutical expenditures and prices have been the focus of a spirited debate, but one that too often is based on incorrect or incomplete information. Our purpose is to restore some balance by providing some of that missing evidence. As health insurance premiums have recently begun to escalate, rising prices of pharmaceuticals have been cited by many managed care executives as the chief reason. Further, a recent Harris Poll found that a majority of physicians believe the increase in pharmaceutical spending can be attributed to rising prices.1 Indeed, the increase in pharmaceutical spending has been much greater than the increase in overall prices. As Figure 1 indicates, in the United States, annual spending increased by 14.2% in 1997 and by 15.7% in 1998, and in 1999 total spending was up 18.8%—more than 6 times the overall rate of inflation.2

What many people fail to realize, however, is that spending and price are not the same thing. Indeed, rising prices are not driving increased spending. As shown in Figure 2, over the past 7 years, price increases typically have accounted for about one fifth of the growth in pharmaceutical spending—22% in 1999.

By far, the largest component of growth has instead been expanding sales of existing products, ie, increased utilization of pharmaceuticals. The use of effective new products that improve or expand treatment options is the second-largest component. The new COX-2 inhibitors, for example, can alleviate arthritis pain but have a different side effect profile than earlier therapies.

As a recent editorial in the Wall Street Journal pointed out, "Problems that had defied medical science for decades or centuries are now giving way to new treatments. That is why most of the recent increases in drug expenditures have been for new therapies, not higher prices."4 So there is no reason to think the pharmaceutical research revolution should lead to spending less money for drugs. Increases are far more likely, because treatment opportunities will continue to expand, just as cheaper computer processing power increased the market for computers far beyond what anyone could have imagined when the microprocessor arrived in the 1970s.

Over the past decade, not only have price increases for existing drugs typically been modest, but prices of newly launched products have often been below the prices of established products that compete in the same therapeutic category (see Figure 3). This fact is in sharp contrast to critics’ allegations that new product prices are always higher than the prices of existing medications; often, that simply isn’t true. Two Pfizer-related examples from well-known and widely used product classes illustrate this fact.

The top panel of Figure 3 reports relative launch prices in the class of new-generation antidepressants. Following the initial entrant early in 1988, subsequent prescription antidepressants were introduced at prices lower than that of the original. The decline in launch prices has been even more dramatic among prescription treatments for erectile dysfunction. As shown in the bottom panel of Figure 3, the most recent entry into this class was priced 61% lower than the original product despite having a treatment modality that many consider to be preferable to those of its predecessors.

When the price and quality of a therapy are both considered rather than price alone, a pharmaceutical innovation often yields a reduction in the total cost of treatment. Beyond that savings, it’s the value––not just the price––of a treatment that matters. And that value comes from improvement in health, not just reduction in costs. Indeed, even if a new product contributes to higher total costs, it may still deliver value for money, since better health and better functioning are worth paying for. Leading economic researchers put it this way: "If medical care increases in cost without much improvement in health, that would be an increase in the cost of living. If medical care increases in cost but the value of that care rises over time, the cost-of-living index would be falling."5 As one example, the cost of antidepressant medications increased on average about 20% from 1991 through 1995, as measured by the official US price indexes. But this measure overstates "real" price increases, according to other economic re-searchers. When quality improvements embodied in the newer drugs are taken into account, the "price of equivalent treatment" fell by between 22% and 32% over the same period of time.6

So yes, pharmaceutical expenditures have risen, and if medical science continues to advance, they will continue to rise.

People want to be healthy, and so they make use of the growing menu of high-quality medicines that treat more medical conditions with increasing effectiveness, convenience, and reduction in side effects.

Prices Vary...

Another common critique of pharmaceutical prices is that they are higher in the United States than elsewhere, and that this shouldn’t be. Blame for this is often laid at the feet of pharmaceutical manufacturers, with the implied "prescription" being that manufacturers, or perhaps governments, should act to equalize prices at a low level everywhere, including in the United States. This view overlooks a number of truths about the pharmaceutical marketplace.

One reality is that price variation is an entirely normal phenomenon. It’s true for prescription pharmaceuticals at the retail level—a level beyond the control of manufacturers. It’s true across countries for all manner of products. It’s not surprising: local supply and demand conditions differ from place to place—not only from country to country but even from street corner to street corner.

...From Drug Store to Drug Store...

A telephone survey of pharmacies in Vermont conducted in 1999 showed substantial variation in the prices different pharmacies charged for the same product, and similar studies elsewhere have shown the same pattern.7,8 As Figure 4 shows, even in this sample from the state of Vermont, the range from highest to lowest price per prescription ran from 20% for an antidepressant prescription to 61% for an antibiotic prescription. This variation reflects a too often overlooked determinant of price variation: different pharmacies set different markups, in turn reflecting local differences in rent, labor, and other costs and differences in profit margins. This important source of price variation is entirely separate from any action by a pharmaceutical manufacturer.

When Internet "drugstores" are included in price comparisons, there is even greater price variation. In December 1999, two Internet pharmacies were selling for $34 the same antihypertensive prescription for which the lowest Vermont survey price was $42.

...and From Country to Country

Within the United States and around the world, prices of all kinds of products and services vary widely. Differences exist not only in the prices of basic consumer items such as food and groceries but also in the prices of items highly dependent on creative ability such as college tuition and professional services. In many international comparisons, prices are highest in the United States, but in many others, prices are higher elsewhere.

Figure 5 shows that a McDonald’s Big Mac® costs more than three times as much in Switzerland as in Hungary, with the United States’ price in the middle of this range. Figure 6 shows that the price of a Starbucks® cappuccino is about twice as high in Tokyo as in New Zealand. And Figure 7 and Figure 8 show how prices vary within the United States for residential electricity, and across European cities for a variety of grocery products.

A Host of Reasons—and a Context for Interpreting Published Price Comparisons

Pharmaceutical prices reflect many of the same forces as prices for other products, but there are also some special reasons that pharmaceutical prices vary from country to country. Unfortunately, too often, colorful news stories seem blind to these facts.

As one example of reports that attract attention but do not provide a complete picture, in 1999, Public Citizen issued a study of international prices for antidepressant and antipsychotic medications,12 in which it was reported that 1997 prices in other developed countries were uniformly and substantially lower than they were in the United States. The study suggested that if our government were to impose the same national health insurance systems that exist abroad, patients would save money and enjoy greater access to medication. Among the products whose prices were examined was one sold by Pfizer. The panel in Figure 9 shows the international prices Public Citizen reported for the 50-mg dose of this product, as well as the prices Pfizer actually charged at the time.

Although there were some discrepancies between the reported prices and the actual prices, the general pattern for the 50-mg dose is consistent with Pfizer’s actual prices. The price was indeed higher in the United States than in the other countries in the sample. But that is not the whole story.

Although this product is available and widely used in both 50- and 100-mg dosage forms, the study did not examine the price of the 100-mg dose. If it had, perhaps Public Citizen’s conclusion would have been different.

Rather than being at the top, the US price for the 100-mg dose was actually in the middle of the distribution of countries, as shown in the panel of Figure 9. In large part, this is due to the fact that in the United States, the price of this product, like that of many medicines, does not vary substantially with the strength of the dose. This allows a physician to choose the dose that best fits a patient’s needs without having to consider the relative cost of different dosages. In most other countries, this is not the case. Regulatory authorities often either encourage or require higher strength dosages to be sold at higher prices. Without investigating the relationship between dose and the international prices of other products in the Public Citizen report, it is not possible to say how broadly this pattern applies. However, it is clear that many such reports on international price comparisons, which often attract media attention, are missing an important part of the story.

In Pfizer’s case, this phenomenon is not limited to one product. Figure 10 shows the prices Pfizer charges for several products sold across the Americas and Europe. Although not every product or dosage is sold in every country, the general pattern is clear: as dosage strength increases, the US price tends to move down the price distribution scale.

This drives home the point that when considering international price comparisons, one must look at the entire picture—not just at a single dosage, a single product, or even a selected set of high-visibility products. Too narrow an analysis can yield an incorrect impression of the real situation.

Exchange Rate Variations
Another factor that leads to price differences across countries is movement in currency exchange rates over time. This is a phenomenon that is clearly beyond the control of pharmaceutical manufacturers.

Consider the hypothetical example in Figure 11: Suppose a product had been first sold in 1980 in 5 countries at the same price ($1), and that a manufacturer had kept its prices in local currencies unchanged in all 5 countries through 1999. Looking at the prices translated into dollars, an observer would see a wide range across the 5 countries develop over time.

Five years after launch, changes in the relative value of the currencies would have made the product appear 45% cheaper in the United Kingdom than in the United States. In 1999, the same product would have appeared to be 28% cheaper in the United Kingdom but over twice as expensive (as in the United States) in Japan.

Since few foreign governments allow pharmaceutical price changes to fully adjust to currency fluctuation, the longer a product has been on the market, the greater is the potential for exchange-rate driven international price variation, regardless of what the manufacturer has done.

Mexico is not shown in Figure 11, but consider the following: Despite Mexico’s recent periods of currency stability, the peso has generally been very weak relative to the dollar. From 1980 to 1990 it lost over 99% of its value, and from 1990 to the present, it has declined by an additional 70%. No wonder people go to Mexico to buy things. And not just pharmaceuticals. When peso prices do not fully adjust to a decline in currency value, US buyers see bargains in a wide variety of products.

Differences in Prices Consumers Pay Reflect Differences in Retail Margins
Even as retailers’ price decisions lead to variations in local pharmacies’ prices in the United States, so do differences in retail margins contribute to international price differences paid by consumers or insurers. For example, although the manufacturer’s price of one erectile dysfunction medication is comparable across Europe, differences in retail margins, which in many countries are set by the government, generate a wide spread in the price actually paid for this product, as Figure 12 shows. While the manufacturer’s prices vary by less than 10% across these countries, retail margins vary by more than 650%, resulting in a 63% range in retail prices.

Different Liability Systems
Yet another contributor to international variations in pharmaceutical prices is the fact that countries’ legal systems differ. The US product liability system is more stringent than in most other countries. The experience of the DPT vaccine for children provides a telling instance. Lawsuits in the 1980s claiming children had been injured by the pertussis (whooping cough) component of the DPT vaccine drove the price of that vaccine up by about 2000% and caused many suppliers to stop selling the vaccine altogether (see Figure 13).

Generics
Dosage form, currency fluctuation, retail margins, liability systems—these are only a handful of the reasons that pharmaceutical prices differ from country to country. Yet another is the extent and nature of the generic drug industry. The United States has a vital generic drug industry, largely as a result of the 1984 passage of the Waxman-Hatch Act. Under this law, the entry of generic products into the marketplace is greatly facilitated when an innovative product’s patent expires, while patent life lost in the clinical testing and regulatory approval processes is partially restored to new drugs approved by the FDA. The effect of generics on the prices American consumers pay for drugs is significant. In this country, generics compete fiercely with each other, with the innovator brand, and with other patented products. Upon patent expiration, generic products enter the marketplace–– driving down the sales of the nonpatent-protected innovator products and driving down the prices of the generic counterparts through market competition. Strong protection of intellectual property coupled with rapid generic entry after patent expiration provides the best balance between the aims of access and innovation. The incentive to develop improved treatments is preserved, and the cost of new treatments falls rapidly after patents expire. As a means of cost containment, the entry of generic therapies is and promises to be increasingly important in this country. As illustrated in Figure 14, in the year 2000 alone, it is estimated that 39 innovative drugs with 1998 worldwide sales exceeding $13 billion will lose patent protection, and the next year, an additional 34 drugs with more than $15 billion in 1998 revenues will be subject to generic entry.16

In other countries, in contrast, the role of generics is less significant, with lower generic market shares and a smaller gap between brand and generic prices leading to less of an effective price-lowering impact.

If you want to look at what the ultimate payers—consumers and taxpayers—are paying for medicines, look beyond what a brand-name manufacturer’s price is, to the average of brand and generic prices at the retail level.

Many other complexities confound cross-country comparisons of pharmaceutical prices. Indeed, to assess the overall impact on consumers, it is essential to look at an entire market basket of products. An extensive body of research by the Wharton School’s Professor Patricia Danzon shows, among other things, that focusing only on the prices of a handful of visible, internationally marketed brand-name products is very likely to overstate the degree of any price disparity.17,18

Placing Price Variation in Context
To get a sense of perspective on what prices in different countries really mean, it helps to look at the cost of goods relative to what has to be given up to buy them. Average individual incomes around the world, even in the developed world, vary substantially. For example, although people often think of Canada as having essentially the same standard of living as the United States, average income in Canada (measured by per capita gross domestic product) is about 30% lower than it is in the United States.19

When viewed in comparison with the hours that must be worked to purchase medicines, international price variations take on a new look. Returning to the examples from the Public Citizen study discussed earlier, Figure 15 shows that on the criterion of hours worked, the 50-mg dose prescription is by far costliest in Mexico—far more costly than in the United States, Canada, or Europe. By this measure, this product in Canada is about as costly as in the United States, and, at least for the 100-mg dose, both countries are among the lowest "cost" countries around. A similar analysis for other drugs would likely show similar patterns.

This perspective sheds light on the folly of seeking price parity between the US and developing countries. Raising prices in such countries to bring them into parity with US prices would place prescription drugs out of reach for many people. On the other hand, lowering US prices to the point of parity with prices in developing countries would seriously harm the pharmaceutical industry’s ability to fund research. Neither of those outcomes is acceptable; they would not provide the greatest benefit for the greatest number of people.

US Pharmaceutical Expenditures in Context

Finally, it is helpful to place US pharmaceutical expenditures in the context of overall spending. Pharmaceuticals are of great value to consumers, of course, and there is no doubt that some consumers are seriously pressed to be able to buy their medications. Pfizer shares the widespread conviction that an appropriate public policy response is needed to deal with this very real problem. At the same time, the magnitude of the problem should not be overstated, especially as a political lever for policy approaches that might well turn out to have serious unintended consequences.

Consider this comparison, illustrated in Figure 16: according to the US Consumer Expenditure Survey, in the course of a year, the average American spends roughly the same fraction of his or her income on pharmaceuticals as on either tobacco or alcohol. Furthermore, as Figure 17 shows, among the elderly, about the same fraction of expenditures goes to pharmaceuticals (3%) as goes toward entertainment.

Even as one looks across income levels, pharmaceutical spending, on average, never exceeds 4% of total spending (Figure 18). This is not to imply that there is no problem. However, as our country considers ways to address this issue, we should bear in mind the true magnitude of the problem. No matter how well intentioned, a policy solution that loses sight of the true need runs the risk of diminishing the promise of future advances in medicine.

As this debate goes forward, policy makers and the public should be aware of the facts regarding pharmaceutical prices, spending, and the value of innovation. Without such an understanding, the legacy we leave to the future may be one of needless illness and lower quality of life.

SOURCES

1. Strategic Health Perspectives. New York: Harris Associates; 1999.

2. Pharmaceutical Pricing Update, Plymouth Meeting, Pa: IMS Health; March 2000:7-3.

3. Bureau of Labor Statistics Internet site. Available at: http:// stats.bls.gov/datahome.htm. Accessed March 6, 2000.

4. Calfee J. Price controls are a prescription for disaster. Wall Street Journal. July 22, 1999

5. Cutler DM, McClellan MB, Newhouse JP. The cost and benefits of intensive treatment for cardiovascular disease. In: Triplett, ed. Measuring the Prices of Medical Treatments. Washington, DC: Brookings Institution Press; 1999:34-71.

6. Berndt ER, Cockburn IM, Griliches Z. Pharmaceutical innovations and market dynamics: tracking effects on price indexes for antidepressant drugs. Brookings Papers on Economic Activity, Microeconomics. 1996:133-199.

7. Sorenson AT. Equilibrium price dispersion in retail markets for prescription drugs. Journal of Political Economy. In press.

8. Von Nostitz G. How seniors can save $$ on drugs: the public advocate’s survey of drugs commonly used by older people. Office of the Public Advocate for New York March 1999. Available at: http://www.pubadvocate.nyc.gov/ agdcdetail.cfm?id=23. Accessed on March 27, 2000.

9. Big MacCurrencies: the Economist offers hot tips on exchange rates. Economist (US ed) April 3, 1999;351:66.

10. Energy Information Administration, Form E1A-826. Available at Internet http://www.eia.coe.gov.cneaf/electricity/epm/epmt55.txt. Accessed on March 27, 2000.

11. Duschenes M, et al. DKBR European Price Survey. United Kingdom: Dresdner Kleinwort Benson Securities Limited; April 1998.

12. Public Citizen report. Available at: http://www.citizen.org/ hrg/publications/1446.htm. Accessed March 6, 2000.

13. Andersson F, McMenamin P. International Price Comparisons of Pharmaceuticals––A Review of Methodological Issues. London, England: Battelle Medical Technology and Policy Research Centre; August 1992.

14. Dow Jones Interactive Quotes & Market Data. New York, NY: Dow Jones & Company; 1999.

15. Manning RL. The revolution in tort law and the market for childhood vaccines. Journal of Law and Economics. 1994; 37:247-275.

16. Generic sales expected to increase starting in 2001. Med Ad News. May 1999; 18(5):46-49.

17. Danzon PM. Price Comparisons for Pharmaceuticals: A review of US and Cross–National Studies. Washington, DC: American Enterprise Institute; 1999.

18. Danzon PM. Pharmaceutical Price Regulation: National Policies Versus Global Interests. Washington, DC; American Enterprise Institute; 1997.

19. World Fact Book 1999. Washington, DC: US Central Intelligence Agency; 1999.

20. Bureau of Labor Statistics Internet site. Consumer Expenditure Survey. Available at: http://stats.bls.gov/csxhome.htm. Accessed March 6, 2000.

Please convey your comments or request additional copies by e-mail to Richard Manning at rmanning@nj.kwsp.com. ©2000 Pfizer Inc. Printed in USA/June 2000. All rights reserved. No part of this publication may be reproduced without written permission. All trademarks are the property of their respective owners. The opinions expressed in this publication represent solely the opinions of the authors and do not reflect the policy or position of Pfizer Inc or of the institutions with which the authors are affiliated, unless this is clearly specified.